Acid phosphatase-1.1, a molecular marker tightly linked to root-knot nematode resistance in tomato

Root knot nematode resistance in tomato is a genetic trait which is determined by a single dominant gene ( Mi ) on chromosome 6 of tomato. Information about the mRNA or protein product is completely lacking, which precludes the cloning of Mi by conventional strategies based on gene expression. However, an acid phosphatase-1 allozyme marker ( Aps-1 ¹ ) is known, which shows tight genetic linkage to the root knot nematode resistance trait. With a view to isolating Mi nucleotide sequences by a positional cloning approach, we have developed a molecular probe for the Aps-1 ¹ gene, using the polymerase chain reaction (PCR) and a pair of primers, corresponding to amino acid sequence information from the protein encoded by the Aps-1 ¹ gene.

The Aps-1 ¹ gene product (APS-1 ¹) was purified from hydroponic tomato roots and tomato suspension cells using conventional low pressure column chromatographic techniques. The most striking purification was achieved by concanavalin A (Con A)-Sepharose 4B affinity column chromatography, which constituted the fourth step in our purification procedure and produced virtually pure APS-1 ¹. Unfortunately, however, this step caused contamination of the APS-1 ¹preparation with Con A released form the column. Therefore, a final Mono Q-FPLC purification step was added to remove the Con A-contamination. The resulting APS-1 ¹preparation was homogeneous in denaturing and non-denaturing polyacrylamide gel electrophoresis. In addition, the purified protein showed co-electrophoresis and co-elution with the APS-1 ¹enzymatic activity in non-denaturing PAGE and gel filtration column chromatography, respectively, thereby demonstrating its APS-1 ¹identity. The yield of our purification protocol was a few μg of APS-1 ¹protein per kg of tomato suspension cells or roots. No major loss of APS-1 ¹activity was observed in any of the purification steps, indicating that the low yield attained is truly reflecting the very low expression level of the Aps-1 ¹gene.

The purified APS-1 ¹preparation was treated with CNBr and trypsin to produce APS-1 ¹peptides. Following purification by HPLC, amino acid sequence analysis of two CNBr and seven trypsin cleavage products revealed 61 residues of APS-1 ¹amino acid sequence.

The amino acid sequence information from two APS-1 ¹peptides consisting of 8 and 14 amino acids respectively, allowed the synthesis of PCR primer pools to direct the amplification of a 2.4 kb Aps-1 ¹ fragment using a genomic DNA template. Crucial for the effectivity of these pools was the limitation of the number of different primers used to account for codon degeneracy. Restriction of the complexity of the primer pools was achieved by incorporating deoxyinosine or the most probable nucleotide(s) at ambiguous codon positions in the 5' part of the primers. On the other hand, efficient primer elongation was assured by including in the primer pool every possible combination of the three 3'-terminal codons. The 2.4 kb fragment was only amplified from a genomic template carrying the Aps-1 ¹ allele and was not found in using a template that carried the Aps-1 ₃ or Aps-1 ¹ + allele. Moreover, the 2.4 kb amplification product was found to reveal RFLPs between a pair of nematode resistant and sensitive nearly isogenic lines, which only differed for the Aps-1 ¹/Mi region. The specific amplification under the direction of the particular Aps-1 allele from which the primers had been derived and its genetic map position provide evidence showing, that the 2.4 kb PCR product represents the Aps- 1 ¹ target fragment. Using cDNA as a template and the same primers that directed the synthesis of the 2.4 kb genomic PCR product, a 490 bp Aps-1 ¹ fragment was obtained. An overlapping Aps-1 ¹ cDNA sequence of 550 bp was amplified using the same upstream primers but a different pool of downstream primers corresponding to a peptide that turned out to represent the C- terminus of APS-1 ¹. The amount of these cDNA-directed amplification products synthesized in 30 cycles of PCR was so low, that their production was only detectable by Southern blot hybridization using the 2.4 kb genomic PCR product as a probe, which provides another demonstration of the very low expression level of the Aps-1 ¹ gene.

In addition to the 2.4 kb genomic Aps-1 ¹ sequence, another PCR product of about 115 bp was obtained using a different pair of primers and either a genomic or a cDNA template. This product was found to comprise several related nucleotide sequences of similar size. Because of the poor performance of the cloned products as a probe, it was not possible to establish their genetic map position by RFLP analysis.

Screening of a cDNA library with the 2.4 kb putative Aps-1 ¹ sequence identified two clones of related nucleotide sequences, one of which apparently represented an Aps-1 ¹ cDNA clone, as it encoded three APS-1 ¹peptides (together 30 amino acids) in the orientation predicted by the PCR results.

'Me nucleotide sequence of this Aps-1 ¹ cDNA clone revealed a stretch of 69 residues of tomato APS-1 ¹amino acid sequence starting from the C-terminus, and showed that peptide VII is the C-terminal tryptic peptide. Another stretch of putative APS-1 ¹amino acid sequence was deduced from the nucleotide sequence adjacent to the upstream primer in the 2.4 kb genomic PCR product, and comprised 56 amino acids sequence information starting with peptide IX at the N-terminal end. The amino acid sequence of tomato APS-1 ¹elucidated so far, did not present major sequence homology with the sequences of any other (acid) phosphatase in the GenBank or EMBL data bases. A striking sequence homology was found, however, with a vegetative storage protein from soybean, VSP-β, that accumulates to very high levels in leaves after depodding of the plants.

The detection of an Aps-1 ¹ -related cDNA clone using the 2.4 kb genomic Aps-1 ¹ sequence as a probe, and furthermore, the sequence heterogeneity among the 115 bp PCR product points to the existence of a family of Aps-1 ¹ -related nucleotide sequences within the tomato genome. Upon comparison of four different nucleotide sequences present in the 115 bp PCR product with the Aps-1 ¹ cDNA sequence, it was found, that one of them corresponded with the Aps-1 ¹ sequence. Neither of the other three, nor the Aps-1 ¹ -related cDNA clone showed a higher homology at the amino acid level with soybean VSP-βthan with tomato APS-1 ¹, which argues against one of these sequences representing the tomato homolog of soybean VSP-β.

In the isolation of an Aps-1 ¹ cDNA clone, we have shown that it is possible to design highly specific degenerate PCR primer pools. Therefore, whenever going from protein to gene, we recommend to try PCR first in order to obtain a highly selective probe, before turning to library screening using degenerate oligonucleotides. Furthermore, since an Aps-1 ¹ cDNA clone is available now, a possible starting point or a useful landmark has been provided for a chromosomal walk towards the nematode resistance gene Mi .